All Times EDT
Keywords: quantitative imaging biomarkers, multidimensional descriptor, phenotype classification, risk prediction, radiomics
Multiparameter quantitative imaging incorporates anatomical, functional, and/or behavioral biomarkers to characterize tissue, detect disease, identify phenotypes, define longitudinal change, or predict an outcome. Multiparameter quantitative imaging biomarkers (mp-QIBs) computed by software are considered software as a medical device (SaMD). The Quantitative Imaging Biomarker Alliance (QIBA) Multiparametric Metrology Working Group developed a series of papers on use cases of multiparameter quantitative imaging: i) multidimensional descriptor, ii) phenotype classification, iii) risk prediction, and iv) analysis of data-driven markers from radiomics. We addressed special considerations associated with each application, compared and contrasted metrics of technical performance, and presented a general framework for estimating and testing technical performance. We concluded with a discussion of the intersection of the clinical utility of an mp-QIB-based decision tool with the technical performance of its output and of the component imaging biomarkers. In this roundtable, we shall share our new developments in mp-QIBs with the audience, open for discussion and seek for suggestions and comments on this line of research.